Journal of Global Positioning Systems (2005)
Vol. 4, No. 1-2: 129-138
Deformation monitoring and analysis using Victorian regional CORS
data
Kefei Zhang1, Youjian Hu1,2, Gangjun Liu1, Falin Wu1 and Rodney Deakin1
1School of Mathematical and Geospatial Sciences, RMIT University , Melbourne, Australia
Tel: [phone], Fax: [phone], Email: [e-mail]
2Department of Surveying Engineering, China University of Geosciences, Wuhan, China
Tel: [phone], Fax: [phone], Email: [e-mail]
Received: 27 November 2004 / Accepted: 12 July 2005
Abstract. This paper investigates the feasibility using
continuously operating reference stations (CORS) in
Victoria (termed GPSnet) for deformation monitoring and
analysis. A number of critical issues associated with the
suitability, geological stability, data quality of the GPS
networks system, the precision and reliability of the
GPSnet solution are investigated using geological
information. Appropriate strategies for GPS data
processing and deformation analysis are investigated. The
absolute and relative displacement of selected GPSnet
stations are analysed using chronological GPS CORS
data and dedicated high precision scientific GPS data
processing software packages. The latest International
Terrestrial Reference Frame is used for deformation
analyses. Detailed data-processing strategies and results
of deformation analyses are presented and some useful
conclusions are drawn.
Results show that the methodology of deformation
analysis and data processing based on the regional CORS
network data is feasible and effective. It is concluded that
high-precision con tinuous tracking data from GPSnet is a
very valuable asset and can provide a technically-
advanced and cost-effective geoscientific infrastructure
for deformation monitoring analysis. By mining the data
from the GPSnet, not only reliable and high precision
deformation information can be potentially obtained, but
also high expenditure required for establishing dedicated
deformation monitoring networks in this area can also be
spared.
Key words: CORS, ITRF, precise ephemeris,
deformation monitoring and analysis
1 Introduction
Natural disasters are of a problem of global concern and
may cause significant social, environmental and human
losses and sometimes, threaten geopolitical stability.
Natural hazards that impact on Australian communities
include earthquakes, landslides, floods, storm surges,
severe winds, bushfires, and tsunamis. In Australia,
natural hazards are estimated at an average annual cost of
$1.25 billion (Geoscience Australia, 2004a). Victoria is
one of the regions where earthquake epicentres are
relatively concentrated. There is potential risk of
earthquakes in this region. For example, Yallourn area
within Victoria is a geologically active part and
experiences earthquake from time to time (Brown, 2002).
Landslide is another considerable geological hazard in
Australia and south-eastern Victoria is a very active
landslide area (see Figure 9). The Earth's surface
deformation due to human activities such as mineral
mining may also cause hazards.
For many applications, such as site selection of important
engineering projects and constructions and their
protection against hazards, the ability to analyse and
predict natural and non-natural hazards is of great
importance. Such ability depends heavily on precise and
reliable deformation information which in turn can be
acquired by using advanced technologies through the
monitoring and analysis of the Earth's surface
displacement, the movement of faults, landslide and some
other deformations .
Due to its high precision, 24 hours availability,
operability under all weather conditions and automation,
GPS technique has been widely used in monitoring
ground movement, deformation and subsidence (He et al.,
2004; King et al., 1995; Kogan et al., 2000). In Victoria,
GPSnet with its high-precision observation data provides

130 Journal of Global Positioning Systems
a technically-advanced and cost-effective geoscientific
infrastructure for deformation monitoring analysis. By
mining the data from the GPSnet, not only reliable and
high precision deformation information can be extracted,
lots of expenditure required for establishing dedicated
deformation monitoring networks in this area can also be
saved.
Methodologies for GPS data processing and deformation
analysis are investigated. The absolute and relative
displacement of selected GPSnet stations subnet are
calculated using chronological GPS data and AUSPOS
(Dawson et al., 2004) on-line scientific data-processing
engine. The feasibility and effectiveness of the
methodologies put forward are discussed and some useful
conlusions are given.
2 Victorian G P Snet
In 1994, Land Victoria, the Department of Sustainability
and Environment (DSE) of the State of Victoria/Australia
foresaw the rapid developments of global navigation
satellite technology and initiated an ambitious project to
establish a set of 20 permanent and continuously
operating GPS Base Stations (GPSnet) across the State.
The primary purpose of the GPSnet is to provide a range
of users with a means of obtaining accurate and
homogenous positioning within Victoria using the space-
borne techno logy. As an integral p art of the new geod etic
strategy for Victoria, GPSnet is being established in
partnership with industry and academia. GPSnet currently
consists of 19 operating base stations and will contain
approximately 24 primary stations upon completion.
Table 1 lists the chronological developments of the
GPSnet stations.
The nominal design spacing of GPSnet stations is
approximately 50km in the Melbourne metropolitan
region and 100km in rural Victoria, but in remote areas
the separations can range up to 200km (see Figure 1). The
Melbourne observatory base station has been connected
to International GPS Service (IGS) network. The network
records, distributes and archives GPS data for accurate
position determination with post-processing techniques.
Seven sites also transmit local real-time kinematic (RTK)
correction signals via radio. The GPSnet system provides
a mechanism for centimetre level positioning relative to
the Australian National Spatial Reference Systems. Real-
time transmission of networked GPSnet data to enable
near instantaneous network RTK positioning services
using a single GPS receiver is currently under
consideration by DSE (Hale, 2004).
GPSnet uses a variety of receivers including Trimble
4000SSE/SSI, 4700 and Leica SR9500 dual-frequency
receivers. The receivers use dual-frequency geodetic
antennas with ground-planes and record C/A code, L1/L2
carrier phase and Doppler data in the RINEX format at all
sites. All antennas are permanently mounted to provide
an uninterrupted view of the surrounding sky. GPS
antennas are usually sited on rooftops of buildings and at
the most stable locations free of multipath. Data
Tab. 1 Chronological developments of the Victorian GPSnet stations
GPSnet stations (date of operation) Year of operation No of Stations (Total)
Ballarat (01/12) 1995 1 (1)
Epsom (01/07) (relocated in 2002)
Melbourne RMIT (01/08) 1996 2 (3)
Geelong (03/09) 1998 1 (4)
Benalla (13/07)
Irymple (relocated in 2003) (26/01) 1999 2 (6)
Colac (30/10)
Mt Buller (19/12) 2000 2 (8)
Swan hill (05/03)
Hamilton (19/03)
Shepparton (06/04)
Walpeup (14/05)
Horsham (02/06)
Yalllourn (21/06) (relocated in 2003)
2001 6 (14)
Cann River (01/09)
Melb obs (IGS station) (18/11) 2002 2 (16)
Clayton (12/02)
Bairnsdale (31/10) 2003 2 (18)
Albury (11/02) 2004 1 (19)

Zhang et al.: Deformation monitoring and analysis using Victorian regional CORS data 131
Fig. 1 Victorian r ural /regional (left) and Me lb o u rne (right) GPSnet base st ati on net wo rk locations and their devel o pment status (Land Victo ria, 2003)
processing is performed in the International Terrestrial
Reference Frame (ITRF) 97 and then transformed to
geocentric datum of Australian (GDA) 1994. Preliminary
results indicate that RMS of daily so lutions is in the ord er
of 2-4mm in easting and northing and 3-8mm in height
using IGS final orbits products (Brown, 2002). Apart
from high-precision geodetic applications, the GPSnet has
been widely used since its inception, including but not
limited to navigation, mapping, GIS, surface deformation
monitoring (eg open pit coal mining), agriculture and
surveying applications (Zhang and Roberts, 2003).
Land Victoria has also developed a mechanism for data
quality check of the GPSnet measurements. This is
measured through visual indications of cycle slips in
carrier phases, multipath effects and data completeness
respectively. Figures 2-4 show cycle slips occurred,
multipath effects and data completeness for Ballarat
station from 7 October to 6 November 2004 (Land
Victoria, 2004).
This information provides a rough idea on the quality of
the data measured in a particular station and this is very
valuable for GPSnet users.
Fig. 2 Cycle slips detected for Ballarat station during October 2004

132 Journal of Global Positioning Systems
Fig. 3 Multipath e ffects d etected for Ballarat station during October 2004
Fig. 4 Data completeness results for Ballarat station during October 2004
3. Method of deformation a na l ysi s
Victorian GPSnet is of high precision (mm level in
horizontal position), and most antennas are of good
quality and high stability. The GPSnet is, therefore,
capable of providing reliable and high-precision
deformation data, such as the determination of both
velocity and direction of Earth's surface displacement,
relative movement of large geological faults, the relation
between the Earth's surface displacement and tectonic
motion, landslide deformation and the crustal
deformation caused by mineral mining.
Figure 5 outlines a detailed process of this investigation
using GPSnet measurements for deformation analysis.
Major steps for data processing, deformation analysis and
some important contributing factors are presented. The
technical requirements and procedures of data-processing
and deformation analysis are usually different for
different types of deformation analyses, and the reference
datum of deformation analysis and the GPS base stations
in the GPSnet should be properly chosen to form an
optimal deformation analysis subnet.
3.1 Selection of deformation analysis datum
A number of ITRFs (i.e. ITRF 93, 94, 96, 97 and 2000)
are involved in Victorian GPSnet data due to historical
evolution. To obtain reliable results of deformation
analyses, coordinate reference frames of GPSnet stations
must be identical. Obviously, the latest and most accurate
reference frame of ITRF2000 should be used as the
unique coordinate reference frame so that both the
position change of entire GPSnet caused by tectonic
motion of the Australian continent and the relative
position changes of GPSnet reference stations caused by
other factors (fault movement，landslide ，mineral
mining，etc.）can be precisely estim ated.

Zhang et al.: Deformation monitoring and analysis using Victorian regional CORS data 133
Fig. 5 Flowchart of GPSnet data processing and stability analysis
Coordinate transformation of an ITRF system to
ITRF2000 can be performed using the transformation
parameters provided by the International GPS Services
(IGS, 2000). The deformation analysis can also be
conducted in GDA94 or Map Grid of Australia (MGA) as
long as the coordinates of GPSnet reference stations in
ITRF2000 are transformed to GDA94 or MGA Grid
using the transformation parameters between ITRF2000
and GDA94/MGA Grid (Dawson, 2002). The
displacement of GPSnet reference stations derived from
the coordinate differences in ITRF2000 from two
different epochs reflects the resultant effects of all
contributing factors on the stability of GPSnet stations. If
the effects from Australian continent motion are
subtracted from the "absolute" displacement, then the
relative displacement of GPSnet reference stations can be
obtained.
To obtain precise relative displacement, it is more
desirable that one relative stable station in GPSnet is used
as the datum of deformation analysis and the GPS
network for displacement analysis is adjusted using a
non-constrained free network adjustment method. In
GPSnet, the "Melbourne Observatory" station in IGS
network should be ideally used as a relatively stable
datum because it is built directly on bedrocks and of high
stability. However, the station was established in 2002
and became operational since November 2002. Before
then, no station in the GPSnet can be regarded of high
stability since all of the GPSnet station antennas are
mounted on rooftops of buildings. Therefore, currently, to
compute and analyse relative displacement of the GPSnet,
relatively stable and precise IGS/ARGN reference
stations close to the GPSnet have to be selected and
subsequently used as a stable datum for relative
displacement analysis of the GPSnet stations, ie the GPS
network for displacement analysis is adjusted using the
free network adjustment method with no fixed datum.
3.2 Formation of deformation analysis subnet
There are a number of different deformation analyses
required, for example, displacement analysis of the entire
GPSnet, local deformation analysis, comprehensive
deformation analysis of the effects of multiple factors and
individual analysis of the effects of a single factor. For
different deformation analysis, GPSnet stations should be
Earthquake, fault
movement, mineral mining,
landslide , di st r i but i on of
base stations, other
geological and geophysical
data, relevant technical
standard,
etc
GPS data collection, quality
check, analysis of precision
and stability of subnet
solution
Normal?
Reference frame transformation,
baseline processing, a djustment
of GPS subnet, precision analysis
and assessmen
t
Selection of deformation analysis
datum, formation of analysis
subnet, GPS data processing
strategy
No
Displacement calculation + significance test
Analyses and conclusions
Antenna stability, solar
activities, Multipath,
satellite status, cycle slip,
base station relocation,
GPS technical
specifications, etc No
Transformation parameters
of ITRF, GDA, MGA grid,
and IGS precise ephemeris,
etc.
Further investigation and
solution
Check and analysis of GPS
data quality (data snooping
and deletion of invalid data,
detection and repair of
cycle slips)
Normal?
Deformation analysi s proc ess

134 Journal of Global Positioning Systems
chosen to form an optimal deformation analysis network -
"subnet" for a parti cul ar correspo n din g purpose.
The subnet used for a certain deformation analysis
purpose should use the same network shape, same
deformation analysis datum and compatible precision
whenever the subnet data is processed and adjusted. By
doing so, potential systematic errors caused by adopting
different deformation analysis datums and minimised.
3.3 Data processing strategy of subnet
Given the fact that the velocity of the Earth's surface
displacement is usually within a few centimetres per year
and the current relative baseline precision of GPS
measurement is in the order of 10-6~10-8, it is, in general,
not necessary to process the GPSnet data continuously or
in a short time interval. Instead, the GPSnet data
processing should be carried out at one time per year or
one time per season scenario (so that the surface
movement/displacement is large enough to be reliably
detected). However, when the Earth's surface is active
due to some reasons, the interval of the data processing
sessions should be properly increased or the session
interval can even be as high as possible in order to extract
real time and kinematic displacement information.
To achieve reliable deformation analysis results, the
solution of the deformation analysis subnet needs to be
precise enough and stable. The precision and stability of
the network solutions are strongly related to the amount
of GPS measurements used to generate the solution,
which is usually measured in the length of observation
time (for a given sampling rate). Research on the amount
of data required has been conducted and solutions from a
minimal of six hours data are usually considered precise
enough and stable for a high precision deformation
monitoring and an alysis (Dawson et al., 2004). However,
there are a number of important factors contributing to
the stability of a GPS network solution, such as the leng th
of observation time, the amount of valid data collected,
baseline length, quality of GPS signal recorded, the
station environment (e.g. multipath, solar activities,
satellite status), and cycle slip, etc. Among these, many
factors vary with time. Therefore, it is necessary to
investigate numerically the proper amount of data
required to generate a reliable and precise solution from
the deformation analysis subnet.
In GPS data processing, analysis of precision and
stability of GPS network solution can be conducted using
precise GPS data processing software, such as GAMIT
(Gamit, 2004), BERNESE (Bernese, 2004) or AUSPOS
(Dawson et al., 2004). A number of trials are carried out
to test the "best" software package for this research and it
is found that all the three packages give very similar
baseline solution. AUSPOS is chosen due to its
automation and access to the solutions in different
reference frames (details see below). Note that for same
deformation analysis network, the same data processing
software should be used whenever the GPS network data
is processed and adjusted so that any potential errors
caused by different computational models and algorithms
can be minimised.
4. Precision and stability of GPSnet solution
The longest baseline (Cann River-Irymple, 723km) in
Victorian GPSnet with a fixed datum derived from three
IGS stations （HOB2，STR1，TIDB）is selected to
form an experimental network for precision and stability
analyses of the GPSnet (see Figure 6). GPS data pre-
processing software "TEQC" (TEQC, 2004) is used for
editing and quality check of the GPS data. The precise
GPS data processing software AUSPOS is used to
generate the solution of the experimental network.
Fig. 6 An experimental network for pre cision analysis of GPSnet
solution (not to scal e )
AUSPOS allows users to submit their data via the
Internet. The RINEX data needs to be static and geodetic
quality (i.e. dual frequency) and the turn-around time of
the processing results is very short. The quality of the
coordinates with 6 hour s of data is: horizon tal precision is
better than 10 mm and vertical precision is better 20 mm
(Dawson, 2002). AUSPOS processing report provides
coordinates in ITRF, GDA94 and MGA Grid, precision
of coordinates, RMS of observations, percentage of
observations removed etc. This information is very useful
for analysing the precision and stability of the
experimental network solution. AUSPOS processing
engine uses IGS precise ephemeris products, Earth
orientation and station coordinate and velocity parameters
and differential technique to several IGS stations. The
data processing is undertaken in accordance with the
International Earth Rotation Service (IERS) computation
standards.
Ir
y
m
p
le
STR1 (IGS)
Cann River
HOB2
(
IGS
)
TIDB (IGS)
IGS station

Zhang et al.: Deformation monitoring and analysis using Victorian regional CORS data 135
The experimental network data recorded on 14 April
2004 is used for precision and stability analysis of
solutions. Figure 7 shows the relation between precision
(sX, sY, sZ) of coordinates computed (in ITRF2000) and
the amount of data used. Figure 8 shows the coordinate
differences (dx, dy, dz) between the coordinate derived
from different session lengths (amount of data) and the
"ground truth" values that are derived from 24-hour data.
From Figures 7 and 8, we can conclude that:
(1) Overall the accuracy of the coordinates derived from
different lengths of observations varies and their
differences can be upto one decimetre level. The
accuracy can be improved when more data is used
and the solution is pretty stable when more than 20
hours of data is used. The differences of coordinates
decrease when the length of the session increases.
This means that solutions converge (to the "ground
truth") when the length of data sessions increases.
(2) When the session length is less than six hours, the
RMS error of coordinates can be more than 15mm,
and the coordinate differences can be more than
20mm, which cannot meet the requirement for a high
precision deformation monitoring. In addition, the
solution is not stable enough, particularly when the
session length is less than 2 hours. Figures 7 (b) and
8 (b) show that some coordinate differences and
coordinate errors derived from 2 hours data are
obviously more than those derived from 1-hour data,
which are, theoretically, not normal. This is most
likely due to the fact that the 2-hour data is too noisy
and there exists a lar ge random erro r (see Figur es 2-4
for example).
(3) When the length of a session is 12 hours, the
coordinate error is about 5mm, and the coordinate
differences can be less than 10mm, which means that
the solution is relatively stable. When the session
length is close to 24 hours, such as more than 20
hours, the precision of coordinates is 3~5mm, and
the coordinate differences can be less than 5mm,
which mean that the solution becomes quite stable.
Fig. 7 Precision evaluation of GPSnet solution in ITR F2 0 00 u si ng different ses s ion lengths of data at stat ions (a) "Cann River"
and (b) "Irymple" respectively
(b) The station "Irymple"
0
10
20
30
40
50
60
70
1 2468 10 12
GPSnet data used (hour)
Coordinate precision (mm)
14 16 18 20 22 24
(a) The station "Cann River"
0
10
20
30
40
50
60
1 2 4 6 8 1012 14 16 182022 24
GPSnet data used (hour)
Coordinate precision (mm)
sx sy sz
dx dy dz
24
-100
-80
-60
-10
-20
0
20
40
60
80
14 26 818 10 14 20
16
12 22
Coordinate difference (mm)
GPSnet data used (hour)
(b) The station "Irymple"
-60
-50
-40
-30
-20
-10
0
10
20
30
2 4 6 8 10 12 14 18 20 2224
Coordinate difference (mm)
1 16
(a) The station "Cann River"
GPSnet data used (hour)
Fig. 8 Differences of the coordinate solutions in ITRF2000 using different session lengths of GPSnet data in com pariso n w i t h t he
solution from 24 hours data

136 Journal of Global Positioning Systems
Fig. 9 Schematic figure of the displacement vector at selected stations and spatial relations between GPSnet statio ns an d ge o l ogical features
Thus it can be seen that daily GPSnet solution (24 hours
of data) is of high precision and sufficient stability. It is,
therefore, possible to use this data for high precise
regional deformation monitoring and analysis.
5. Calculation and analysis of deform ation
Figure 9 shows the distribution of some geological
features (earthquake epicentres, faults and landslides) in
Victoria and the spatial position relations between
Victorian GPSnet stations and these geological features.
There are more than 10 relatively large faults within
Victoria and some stations are close to faults
and/landslide sites (eg Epson). Victoria, in particular
south-eastern Victoria, is one of the regions where both
earthquake epicentres and landslide sites are relatively
concentrated. There are potential risks of earthquake and
landslide in this area. In addition, human activities such
as mineral mining can also cause deformation of the
Earth's surface. Therefore, according to the result of
displacement analysis of GPSnet stations, the regional
deformation of the Earth's surface and the stability of
some faults and landslide sites can be inferred.
5.1 Calculation of G P S net Station Dis placemen t
A number of factors are taken into consideration when
choosing experimental network, length of sessions and
epochs of comparisons. These factors include data file
losing, improper data format and relocation of some
stations. GPSnet data from 14 April 2002 to 14 April
2004 and seven base stations (see Figure 10) are used in
this paper for local deformation analysis.
Fig. 10 A seven-statio n G PS n e t s u bn et selected for displac e ment
analysis (not to scale)
There are 21 simultaneous observation baselines in the
subnet. The longest baseline length (Walpeup-Melbourne)
is 399km and the shortest baseline length (Colac-Ballarat)
is 90km. The subnet is adjusted using the free network
adjustment method (with no fixed datum). The absolute
displacements in horizontal directions (∆E=Easting,
∆N=Northing) and vertical direction (∆U=up direction) of
the subnet stations (derived from the transformation of
ITRF2000 to Australian Map Grid) are shown in Table 2.
Walpeup
Horsham
Colac
Hamilton Melbourne
Ballarat
Swan Hill

Zhang et al.: Deformation monitoring and analysis using Victorian regional CORS data 137
Tab. 2 Absolute and re l a tive displaceme n t s o f t he G P S n e t su b n e t s t a ti o n s
Absolute displacement (mm) and velocity (mm/yr) relative horizontal displacement
(mm)
station
∆E ∆N ∆U V V/2
significance
test ∆Er ∆Nr significance
test
Melbourne 21 124 32 130 65 9 -6 -5 ×
Ballarat 24 121 25 125 62 9 -3 -8 ×
Colac 32 125 18 128 64 9 5 -4 ×
Hamilton 35 128 16 134 67 9 8 -1 ×
Horsham 20 135 34 139 70 9 -7 6 ×
Walpeup 33 138 47 148 74 9 7 9 ×
Swan Hill 27 131 45 142 71 9 0 2 ×
The total displacement magnitude "V" is calculated by the
following formula:2 )(
2
)(
2
)( UNEV ∆+∆+∆= . "V/2" is
the mean annual velocity of the displacement. "∆Er" and
"∆Nr" are relative horizontal displacements and are free
from the systematic horizontal displacement of the whole
subnet.
5.2 Deformation Analysi s
The significance of both absolute and relative
displacements in Table 2 are tested. The displacement
significance of the whole subnet is tested using F-Test and
the displacement significance of single station is tested
using T-test. The results of significance test are listed in
Table 2. The symbols "9" and "×" indicate significant and
insignificant respectively. The results of significance test
show that the absolute displacements of all the subnet
points are significant. The absolute displacement directions
of all the subnet points are shown in Figure 11. The
average displacement velocity of the subnet points is 6.8
cm/year. Both the magnitude and direction of the absolute
displacement of all the base stations in the subnet agree
well with the velocity of approximately 7cm/year and
direction of current Australia tectonic motion (see Figure
12) derived from other IGS measurements (Geoscience
Australia, 2004b).
Since the precision of vertical coordinate (height) is about
2-3 times lower than that of horizontal coordinates, the
relative vertical displacements of the subnet points are not
precise enough and reliable for high precise deformation
analysis. Therefore, the relative vertical displacement of
the subnet is not analysed in this paper. The results of
significance test show that the relative horizontal
displacements of all the subnet points are not significant.
Thus it can be seen that the relative h orizontal p os itions of
the subnet points are not notably affected from local
geological features. According to this, it can be inferred
that currently, the faults and/or landslide body near these
base stations are relatively stable. Of course, the stability
Walpeup
Horsham
Colac
Hamilton Melbourne
Ballarat
N
Swan Hill
50mm/yr
Fig. 11 Amplitude and directi on of the absolute
50mm/yr
Victoria
Fig. 12 Australian tectonic motion vector from IGS

138 Journal of Global Positioning Systems
of the faults and landslide body still needs to be
continuously analysed in th e future.
5. Conclusive remarks
The solution of Victorian regional CORS network is not
precise enough and stable for high precise deformation
monitoring and analysis if the amount of data used to
generate the solution is less than 12 hours. The precision of
3D coordinates derived from daily GPSnet solution (24-
hour data) is 3~5mm and the solution is quite stable. This
can meet the requirements of high precision deformation
analysis. Therefor e, continuous tracking data fro m GPSnet
is a very valuable asset and can provide a technically-
advanced and cost-effective geoscientific infrastructure for
regional deformation monitoring and analysis.
The average velocity of the displacement at subnet points
is 6.8 cm/year. Both the magnitude and direction of the
whole subnet displacement agree well with the velocity of
~7cm/year and direction of current Australian continent
derived independently. The relative horizontal positions of
the subnet points are not notably affected from local
geological features. It can be inferred that the faults and/or
landslide body near these base stations are relatively
stable.
Preliminary results indicate that the methodology of data
processing and deformation analysis based on CORS is
feasible and effective. However, further investigation is
required when more GPSnet data covering a larger
chronological span and more GPSnet base stations can be
used for deformation analysis. It is recommended that
geological information needs to be taken into account
when any new CORS stations are established. The
improvement of data quality, stability of antenna,
precision and reliability of the GPSnet solution will be of
great help in the analysis of both absolute and relative
displacement of the GPSnet stations. It is, therefore,
anticipated that the GPSnet will play an important role in
the regional deformation monitoring and analysis.
Acknowledgements:
The authors would like to thank P. Ramm, M. Hale, P. Oates, J.
Millner and E. Retimana from the Department of Sustainability
and Environment, Victoria for their consistent support and access
to the GPSnet data. Financial support ("211" academic exchange
fund program) awarded to Prof Youjian Hu from China
University of Geosciences is gratefully acknowledged. Partial
financial support from the Australia Research Council Linkage
project (LP0455170) endorsed to a research consortium led by
A/Prof Kefei Zhang is highly appreciated.
References
Bernese (2004): Bernese GPS Data Processing Software
Package, Version 4.2, http://www.bernese.unibe.ch/.
Brown N. (2002): Quality Control and Integrity Monitoring of
the Victorian GPS Reference Station Network, FIG XXII
International Congress, Washington DC, April, 2002. 19-26.
Dawson J.; Steed J. (2002): International Terrestrial Reference
Frame to GDA94 Coordinate Transformations,
http://www.ga.gov.au/image_cache/GA3795.pdf.
Dawson J.; Govind R.; Manning, J. (2004): The AUSLIG Online
GPS Processing System (AUSPOS), http://
www.ga.gov.au/nmd/geodesy/sgc/wwwgps/pdf/auspos.pdf.
GAMIT (2004): GAMIT-GLOBK GPS Data Processing
Software Package, http://www-gpsg.mit.edu/~simon
/gtgk/index.htm.
Geoscience Australia (2004a): National Datasets online GIS.
http://www.agso.gov.au/map/national/
Geoscience Australia (2004b): Australia Space Geodesy analysis
Center, http://www.ga.gov.au/nmd/geodesy/sgc/gps/
index.htm.
Geoscience Australia (2004c): Differences between coordinate
systems, http://www.ga.gov.au/nmd/geodesy/datums/
cosys.jsp.
Geoscience Australia (2004d): AUSPOS-Online GPS Processing
Service, http://www.ga.gov.au/nmd/geodesy/sgc/wwwgps/
Hale, M. (2004): Personal Communication.
He XF.; Guang Y.; Ding XL.; Chen YQ (2004): Application and
evaluation of a GPS-multi-antenna system for dam
deformation monitoring. Earth Planets and Space 56 (11):
1035-1039.
IGS (2000): International GPS Services (IGS) - International
Terrestrial Reference Frame (ITRF), http://itrf.ensg.ign.fr/.
King, N. E.; Svarc J. L.; Fogleman E. B.; Gross W. K.; Clark K.
W.; Hamilton G. D.; Stiffler C. H.;Sutton J. M. (1995):
Continuous GPS observations across the Hayward fault,
California, 1991-1994, J. Geophys. Res., 100, 20,271-
20,283, 1995.
Kogan M. G., Steblov G. M., King R. W., Herring T. A., Frolov
D. I., Egorov S. G., Levin V. Y., Lerner-Lam A., Jones A.
(2000): Geodetic constraints on the rigidity and relative
motion of Eurasia and North America, Geophys. Res. Lett.,
27, 2041-2044, 2000.
Land Victoria (2003): GPSnet GPS Base Station Logon and Site
Details, http://www.land.vic.gov.au/.
TEQC (2004): TEQC: The Toolkit for GPS/GLONASS/SBAS
Data, http://www.unavco.org/facility/software/teqc/
teqc.html.
Zhang K.; Roberts C. (2003): Recent Development of the
Victorian High Precision Permanent GPS Tracking
Network with Real-time Positioning Capacity, Proceedings
of the 5th International Symposium on Satellite Navigation
Technology and Application, Melbourne, Australia, 22-25
July.
___________